need. >> put -- put aside whether this was necessary and whether it was an appropriate last resort in a quest for diversity and critical mass, because grutter's not without limits. but i'll put that aside and let me come directly to your question. first of all, if you think about narrow tailoring, you can't tailor to the unknown. if you have no range of evaluation, if you have no understanding of what critical mass means, you can't tailor to it. >> so you have to set a quota for critical mass? >> no. there's a huge difference, and it's an important one that is not well put out by the university of texas. having a range, a view as to what would be an appropriate level of comfort, critical mass, as defined in grutter, allows you to evaluate where you are -- >> so we won't call it a quota; we'll call it a goal, something grutter said you shouldn't have. >> well, justice sotomayor, i think it's very important to distinguish between the operative use of that range, in other words, that's where we are, and we're going to use race until we get there every year in consideration of each application, which was a problem.